Statistics Question

vin17 · #1 01-21-2007, 10:25 PM

I have a range of 1,130,000(lowest value is 120,000 the highest is 1,250,000). I am trying to create 8 mutually exclusive classes with the same class width. For some reason I cannot get the last value of the last class to be exactly 1.25m when using the same width across all classes. Is there not an equation for this?

Help is appreciated.

vin17 · #2 01-21-2007, 10:50 PM

MBH, Ryan, if this could temporarily be x-posted that'd be great.

T50_Omaha8 · #3 01-22-2007, 02:03 AM

I'm not sure if this is any help at all, but if (range + 1) / (the number of classes) does not equal an integer, it should be impossible to fulfill these criteria exactly.

It's pretty damn trivial though.

bav · #4 01-22-2007, 08:00 AM

Looks like you actually have 1,130,001 in the range if it's [120,000-1,250,000]. 8 is not a prime factor of 1,130,001. Make it [120,000-1,250,000) (so 1,250,000 isn't actually allowed) and you have 1,130,000 in the range which would be 141,250 per class.

vin17 · #5 01-22-2007, 10:07 AM

Ok, well forget 8 classes. I can have anywhere from 6-15. What I need is to be be able for my first class to be [120,000 - X] and my last class to be [X - 1,250,000]. The number at the end of each class range cannot equal the the number at the start of the next range.

bav, I need all the numbers from [120,000 - 1,250,000] inclusive. If needed, more classes could be added as along as the lowest range value is the same difference from 120,000 that the highest range value is from 1.25m.

edit: it'd be nice if the sets of ranges started and ended in either ,000/,999 or ,001/,000

Thanks.